Time delay

Jim Thompson a écrit :

Make that 2/Fo.

At Fo, the delay is cut by half. If you want 5% delay variation you'll have to stay below Fo/5. For 10% just stay below Fo/2.

So you were right, but sort of the other way :-)

For 10% delay variation the OP will need at least 40KHz Fo, which will give 50uS per section and will require 20 sections.

Fred Bartoli a écrit :

Oops. I meant: For 20% just stay below Fo/2. BTW, 10% variation is for Fo/3.

In message , dated Fri, 8 Sep 2006, Adam writes

I can't understand what you are trying to do. 10 ms is a path difference of only 10 feet. You don't need to be concerned about delay for distances under 35 feet, because the ear integrates all signals received in that time-frame and simply doesn't hear the delay.

Is it that you have a line of 7 loudspeakers 10 feet apart? If so, there are ways of preventing interference from distant sources that don't use delay and cost nothing.

Please give much more data about what you are trying to do.

I just saw a whole bunch of stuff come up. Some of these could have been bucket brigade. One mentioned 250 ms delay in the heading. Not sure if anybody actually still manufacturs these. I suspect a CPU or DSP and a lot of memory will be cheaper.

Tam

Some of us still own a set of "sheet metal drill bits", though I've had clerks tell me, from time to time, that no such thing ever existed.

"sheet metal drill bits" have a small center pilot point and sharp "ears" that actually cut out a small circle of sheet metal, rather than the mangled mess you get when you try to drill sheet metal with a conventional "steel" bit.

...Jim Thompson

In message , dated Sat, 9 Sep 2006, Jim Thompson writes

They are sold for drilling wood now.

In message , dated Sat, 9 Sep 2006, John Fields writes

We have stepped bits like those, called 'Bradrads', and conical ones without steps called 'Conecuts'.

I've not seen 'Bradrads', but I've seen 'Conecuts'.

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

I didn't remember the range well, but what I had in mind is shown at...

Back when I was a kid I used multiple sections of such a scheme to trap record play-back "pops" when I was transferring to tape.

...Jim Thompson

Digital is the way to go, you could use a small PIC (1 for each channel). A PIC18f252 would easily give you up to 35mS.

Here